Metadata-Version: 2.2
Name: doraemon-torch
Version: 0.0.3a0
Summary: Doraemon
Home-page: https://github.com/wuji3/Doraemon
Author: duke
Author-email: dk812821001@163.com
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torchmetrics==0.11.4
Requires-Dist: opencv-python==4.7.0.72
Requires-Dist: numpy==1.24.3
Requires-Dist: tqdm==4.66.4
Requires-Dist: Pillow==9.4.0
Requires-Dist: grad-cam==1.4.8
Requires-Dist: timm==0.9.16
Requires-Dist: tensorboard==2.16.2
Requires-Dist: prettytable==3.10.0
Requires-Dist: datasets==2.20.0
Requires-Dist: imagehash==4.3.1
Requires-Dist: transformers==4.48.3
Requires-Dist: torch==2.5.1
Requires-Dist: torchvision==0.20.1
Requires-Dist: torchaudio==2.5.1
Requires-Dist: faiss-cpu==1.7.2
Dynamic: author
Dynamic: author-email
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# <div align="center">DORAEMON: Deep Object Recognition And Embedding Model Of Networks</div>

<p align="center">
<img src="./misc/doraemon.jpg">
</p>

<p align="center">
<img src="https://img.shields.io/badge/python-3.10-blue.svg">
<img src="https://img.shields.io/badge/pytorch-2.0+-orange.svg">
<img src="https://img.shields.io/badge/torchmetrics-0.11.4-green.svg">
<img src="https://img.shields.io/badge/timm-0.9.16-red.svg">
<img src="https://img.shields.io/badge/opencv-4.7.0-lightgrey.svg">
<a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue.svg"></a>
</p>

## 🚀 Quick Start

<summary><b>Installation Guide</b></summary>

```bash
# Create and activate environment
python -m venv doraemon
source doraemon/bin/activate

# Install Doraemon
pip install doraemon-torch
```

## 📢 What's New

- **[Oct. 2024]** [Content-Based Image Retrieval(CBIR)](doraemon/models/representation/README_CBIR.md) support added with ConvNext backbone
- **[Apr. 2024]** [Face Recognition Task(FRT)] launched with various backbones and loss functions
- **[Jun. 2023]** [Image Classification Task(ICT)](doraemon/models/classifier/README.md) released with advanced training strategies
- **[May. 2023]** Initial release of Doraemon

## 🎯 Implemented Methods

|Category | Methods |
|----------|---------|
| Optimization | SAM, Progressive Learning, OHEM, Focal Loss, Cosine Annealing |
| Regularization | Label Smoothing, Mixup, CutOut |
| Attention & Visualization | Attention Pool, GradCAM |
| Representation Learning | ArcFace, CircleLoss, MegFace, MV Softmax |

## 🔮 Supported Models
 
**Doraemon** now supports 1000+ models through integration with Timm:
 
- All models from `timm.list_models(pretrained=True)`
- Including CLIP, SigLIP, DeiT, BEiT, MAE, EVA, DINO and more

[Model Performance Benchmarks](https://github.com/huggingface/pytorch-image-models/tree/main/results) can help you select the most suitable model by comparing:
- Inference speed
- Training efficiency 
- Accuracy across different datasets
- Parameter count vs performance trade-offs

> For detailed benchmark results, see [@huggingface/pytorch-image-models#1933](https://github.com/huggingface/pytorch-image-models/issues/1933)
